Skip to: Navigation | Content | Sidebar | Footer


Full Archives

New Server

October 31

Welcome to the new server! I've been running a few other sites on Dreamhost for the past six months or so and have been generally happy with the service. They keep raising the monthly bandwidth and storage space, so barring a Slashdotting or two, I'm pretty confident they'll be able to keep up with my needs. I'm not totally thrilled with their oft-praised control panel, it seems a bit slow and quirky. Even after a few months of using it, I'm still never sure quite where to go when I need to do something specific.

But anyway, I'm enabling comments to make sure those still work. I crammed in an upgrade to MT 3.2 while I was setting things up, and it seems relatively stable, but I was forced to make a change I'm not completely sure about yet. Hold off on commenting for now; looks like the templates are a bit mangled at the moment. Okay, comments are a go again. Not perfect, but they should at least be functional.

Permalink › | 26 comments

Changing Web Servers

October 25

I'll soon be going through the almost-always-painful process of switching hosts on a site I'm working on. I've become much better at planning for these over the years, but the complexity of sites I build has added more variables to the mix that I need to account for which probably negates any advantage I thought I had.

There is still an uncomfortably large amount of guesswork and finger-crossing leading up to the switch. Inevitably, some setting or other doesn't get transferred which makes for a day or two of debugging after the fact.

Here's my working checklist of things that need to happen. Any you'd add?

Transfer files.
All the .html, .php, .jpg, etc. files sitting on the old server need to move to the new server. In the past I'd have used an FTP program to take care of the job, but thanks to rsync I can much more quickly mirror the two servers. At the moment I transfer the data from the old server to my local hard drive temporarily, and then re-upload to the new server. I suspect I can skip the middle step of my local system and do a direct transfer, but it never hurts to have the data in more than one spot anyway.
Transfer databases.
All MySQL databases need to move as well; though I'm generally not a big fan of it, PHPMyAdmin is a bit easier to wrangle than the command line (sometimes), and has a one-click import/export function.
Change absolute paths.
Some of my custom PHP relies on absolute server paths for include files and the like file output. As much as possible I've been declaring the path as a constant in global file, so it makes for a somewhat easier migration; except that the path to the global file itself is usually absolute, so unfortunately that does mean a bunch of find and replacing.
Verify permissions.
I've been using PHP's ability to write files to the server in a function that grabs EXIF metadata from images and caches that in static text files. Problem being, PHP's file management honours file permissions, and I've yet to de-tangle which combination of permission/ownership I need to assign the script and write directory in order to securely write these files; to actually make it work the way I'd expect, I've simply assigned 777 to the directory and treat the cached data as disposable. There must be a better way, but I haven't found it. The upshot for the migration is namely that I need to ensure these permissions remain unchanged. That's the other advantage to using rsync over FTP, it should do the work for me.
Verify passwords.
I'm hoping the directories I've password-protected using .htaccess and .htpasswd files are using relative paths. But in case not, I'll need to check if the directories that need to be protected are such.

Luckily I haven't installed any PHP modules or done any extensive customization of Apache, so assuming the new host has a similar setup, that shouldn't become a problem.

And I have a static IP on the new server, which allows me to test ahead of time, instead of waiting for DNS to propagate before I go into mad panic mode to fix the things that get busted from the switch.

The main question mark for me is mail; assuming I set up the user accounts ahead of time, the transfer shouldn't interrupt regular mail delivery (aside from the switch in server settings). But I'm not nearly well-versed enough in mail software to know what'll happen for sure until after I throw the switch.

Again, any glaring omissions would be more than welcomed. I'm hoping I've accounted for all I need to, but general advice is always appreciated.

Permalink › | 24 comments

CSS Validator Bug

October 25

If, like me, you tend to drop units when using the CSS line-height property, like so:

h2 {
   line-height: 2;
}

Then you may have noticed all of a sudden that the validator is reporting this as an error. The validator is, in fact, wrong. It has been reported. There's a workaround. It'll be fixed soon. Don't bother working around. That's all.

Permalink › | no comments

Design Maturity

October 21

Via Digital Web, Jess McMullin has come up with one of the best summaries I've seen of what design can be. Make sure to grab the continuum model PDF.

Briefly summarized, the PDF describes design as a continuum that progresses from no conscious design to focus on style, then to the form and function level, eventually settling into problem solving, and finally, if you're lucky, morphing into something redefining and potentially disruptive.

This almost perfectly describes my own professional development, perhaps you've experienced similar. Once upon a time, I was all about style. I hung around DiK and K10k and the like. Then I started solving more complex UI problems and working with CSS, which put me into the throes of the form and function phase. But simply placing elements on a page is getting old; I'm interested in influencing what those elements are in the first place, and how they work together.

All along I've seen shades of big picture problem solving in my work, so I don't think you're ever strictly confined to one level; and just because you're a problem solver doesn't mean style is 'beneath' you. That's the one issue I'd take with the PDF as-is; the top-to-bottom layout implies a hierarchy, placing stylists at the bottom. Some days it's my job to simply style, and that's not necessarily a bad thing.

Permalink › | 14 comments

Digital Camera Stuff

October 19

A few items I've found recently related to digital cameras and digital photography. (In case you were wondering what I ended up doing after looking for SLR pointers in the summer, wonder no more.)

Aperture
Apple announced some (expected) hardware updates today, but from the left field comes news of Apple's new high-end photo processing software. Looking through the product overview and videos, it's definitely not a Photoshop competitor for most people. There are overlapping features, but it feels far more like a complementary tool than a replacement.
CCD Flaws
Own a point and shoot digital camera from Canon, Sony, Konica Minolta, or Fuji? Bad news; it might be defective.
Why Focus-Recompose Sucks
After discovering that the middle auto-focus point on most SLRs is the most sensitive, I wondered why I needed to bother with the rest of them? Why not just focus-recompose every time? And so that's what I did. And now thanks to this article, I realize why my last month of shooting has been abnormally out of focus. Oh well.

Permalink › | 15 comments

Vancouver CSS Job

October 19

Here's a quickie for anyone in town who might be looking for work.

Local web shop Combustion Labs is looking for a CSS developer. The pay is good, the office is beautiful, the team is smart (I ran an afternoon workshop earlier this year that saw a lot of polite nodding and smiling as I went through material that they had learned ages ago) and the benefits are top notch. From their help wanted posting:

Do your influences include Apple, Zeldman, 37 Signals and Stop Design? If so, you have found your new home.

Not to mention the daily catered lunch program. Need I say more? Check out the craigslist ad. (Please direct all inquiries their way. I'm not being paid to post this, I'm curious to see what sort of local response it will generate.)

Permalink › | no comments

Type Resources

October 18

Vitaly Friedman is collecting license-free fonts. That's right, free as in beer; you're able to use these for commercial work, and unlike most free font collections, there are some in here you may actually want to use. Consider plugging a few of these into your new, free copy of Font Explorer.

Yanone Kaffeesatz
One of the more original faces in the set, the large x-height makes it more appropriate for headlines than body copy. Four weights are available, for a good variety of choice.
Piqiarniq
Reminiscient of Gill Sans, and probably notable more for its origins than its aesthetics. Piqiarniq was commissioned by the government of Nunavut, a territory in the Canadian arctic to provide characters for the Inuktitut language.
Day Roman
A baroque face with more than a passing resemblance to Garamond—compare them side by side below the fold on this page. Plenty of ligatures to keep typofetishists happy.
...and 17 more.

Caveat emptor though. Babelfish seems to indicate that the updated note on the Day Roman page casts doubt on the origins of the font, with the word 'plagiarisms' jumping out in particular. As is the case with any resource available for free on the internet, it can't be a bad idea to research the origins yourself before committing to using them in your commercial work.

And hey, nothing's wrong with paying for your type. Even if you're not looking to shell out a few hundred for an OpenType family with a dozen variants from a major foundry, there are plenty of independent font designers out there.

Mark Simonson's new Proxima Nova is off to a great start, but his back catalogue is not to be missed either. P-Type Publications has released a couple of Indie Fonts collections, which feature samplings from your favourite independent foundaries like Test Pilot Collective and P22. And for the whimsical, don't miss Robot Johnny and Larabie. (I've had Robot Johnny's 'Girls Are Weird' installed for, like, years and years.)

Update: Caveat indeed. Day Roman aside, other fonts in the 'license-free' collection actually impose non-commercial licenses, meaning you can not use them without paying for them, so use the list with discretion. Again, make darn sure you know the origin of the 'free' resource you found online before using it in your own work.

Permalink › | 9 comments

Firefox Growth

October 18

Why has Firefox growth slowed? I received a similar question in my inbox the other day, followed by this tidbit:

Despite all the publicity online and even big-media coverage, I'm still amazed at how many people I meet who have not the faintest idea Firefox exists. Even 20somethings at my office—a dot-com, fer cryin' out loud. They just don't know it's out there.

It's a good question, but you don't have to look far to find the answer. Here's an analogy: I have no idea what the latest and greatest is when it comes to server hardware; I'm in the industry, so maybe I should, but I just don't care that much because it doesn't generally affect me.

Last year I predicted that Firefox would have a 25% share of the market. Obviously I was off a touch, as I'd guess the actual numbers are only a bit more than half that at the moment. Still up from where they were when I made the prediction, but I was a bit optimistic. Spreadfirefox.com reports almost a hundred million downloads now, but big picture trends seem to indicate that Firefox has achieved maybe 15% of the market share amongst actual users, give or take 5%.

I figured the ongoing Windows security issues would continue to drive Firefox growth. But Microsoft has continued pushing out security updates, and let's be honest, PCs are cheap enough to be disposable to a lot of consumers who can't be bothered to fix them; a new PC with Windows XPSP2 will only run you a few hundred dollars. Beats paying $70 an hour for someone to remove spyware from your old one.

Still, double the market share in a year isn't bad. 25% may yet be achievable, and if so, it makes Firefox a large enough force that your average non-standards aware web developer can no longer justify coding for IE-only. And that's a start.

Okay, and since we're talking web site statistics, I guess I need to clarify: yep, Site X probably does have a higher percentage of Firefox users. Stats vary from site to site; this one sees almost five times the number of Firefox users as it does Internet Explorer users for example. But your audience likely doesn't represent the global market share, and that's what we're talking about here.

On the other hand, if you want to tell me the distribution of browsers coming into your site off search engine queries, that's a far more interesting number.

Permalink › | 38 comments

The Local Web

October 17

Something I've been thinking about lately is the idea of localized web sites, or sites restricted to a smaller audience by either content matter or technological barriers. Having a global audience is great for meeting new people and being exposed to ideas and mindsets one may not have been aware of, but sometimes that's too much exposure.

For example, as more members of my family slowly become wired, the idea of having a common, family-only place for us all to go and talk and post pictures seems like the next logical step beyond email. I'd actually prefer that to email, given shared common archives and search options, but it might be a barrier for other members of the family who aren't as comfortable with technology.

So it's been interesting watching details emerge about Project Comet, a new product from Six Apart. It's looking as if it might just be exactly what I think our family needs, with features like photo sharing, media lists, and private weblogs. What I'm really interested in seeing is the content creation interface, and how they plan on making it accessible and user-friendly to people who aren't technically proficient.

Also picking up in my neck of the woods are community-oriented sites and weblogs. Locally focused sites with content only relevant to people within a certain geographic region are a niche that hasn't traditionally been done well on the internet. It seems to me that when you have a company driving the local content creation, there has to be a return in order to justify the resources, so you end up with company directory after company directory, maybe some events listings, a bit of weather, and not much else. Craigslist is a notable exception which has been expanding like crazy, but unless you live in one of the earlier cities it added, it's not quite there yet. (Although it was good enough to find us a new place to live last time around, replacing the for-pay listing service I'd used prior.)

But in the hands of passionate individuals who simply want to share their knowledge and enthusiasm for the place they live, it's a totally different experience. I've enjoyed watching new local content launch at an increasing pace, mostly in the form of weblogs thus far.

Long-running local favourite VanEats is written by enthusiastic technologist and food critique Roland Tanglao, a name some might recognize from the Bryght team. Also run by the same folks is UrbanVancouver, a sort of local news aggregator pulling from numerous sources around town. (Disclosure: I had a hand in the design a few years back.) Local duo Ianiv and Arieanna run Vancouver Coffee, a subject close to my heart. Neighbourhood-specific blogs kitsilano.ca and Beyond Robson just kicked off this spring, and even the local businesses are getting into the spirit: check out the amount of detail and obvious love for the location that has gone into this site for a relatively small commercial corner of the city at Macdonald and Fourth. The streetscape photo tours are similar to a feature amazon.com launched a while ago, but never found its way north of the border; I wish all commercial streets up here had online tours like these.

Permalink › | 19 comments

Spam | October 16

There are just so many things about this that are wrong, I don't even know where to begin.

View on Flickr › | 5 comments

Internet Control Struggle

October 14

If you've been following the news, you may be aware of the growing struggle over control of the internet. At issue: the United States wants to retain control, no one else wants to let them.

A real fear is that if the major players can't come to some sort of agreement, it could lead to country-specific segregation. The global nature of the net would crumble as national sandboxes prevent incoming or outgoing connections to other local networks, through active censorship or simply benign incompatibility.

But I guess there's a silver lining. With ICANN's moves of late being the approval of boneheaded top-level domains like .mobi, among others, I wouldn't be terribly upset were they replaced.

Permalink › | 29 comments

Office UI

October 14

There's a great new blog over at MSDN from Jensen Harris, a member of the Microsoft Office user experience team. In case you haven't seen the screenshots floating around, the forthcoming Office 12 from Microsoft is undergoing a pretty significant UI update, so reading one of the people working on it has been fascinating.

Jensen is writing in detail about historical UI decisions, evolution of the office suite, and some user testing results that might surprise you. A lot of their findings are definitely not Office-specific. Some notable recent posts:

Stroking the Keys in Office 12
In which Office visibly exposes shortcut keys when you hold down alt. Great idea; I believe Windows has been using a similar concept with underlined shortcut keys in menus for years, but this is a much more obvious way of emphasizing the connection between shortcut and function. Now if one of the major browsers were to get accesskeys working this way...
Be Willing To Be Wrong
In which Office's new Ribbon toolbar tried and failed as a left-to-right layout in order of descending importance. Turns out items near the middle are far more visible. This is my favourite post in the bunch so far.
Ye Olde Museum Of Office Past
In which Office's UI history is recounted from the 80's up to the last major release.
Saddle Up to the MiniBar
In which Office implements a mini formatting menu on selection. I love the idea, but from the comments on the post it sounds like it's tightly saddled to text formatting only. A shame, I'd have expected it to be modal, taking it's specific palette of tools at any given time from the Ribbon toolbar.

I have to admit, "Microsoft" and "Great UI" are two phrases that don't often appear together in my mind, but credit where it's due. The Office team seems to be doing a bang-up job on the next release.

Permalink › | 11 comments

Bye Bye Tan Hack

October 13

I've been using the Tan Hack for ages, otherwise known as * html, since it's proven a simple and effective mechanism for combatting some of the worst rendering bugs in IE. With only a few similar alternate ways of doing the same thing available, for the past few years it has been pretty much a choice of personal preference whether you'd go with Tan or conditional comments or html > body.

At the same time, I suppose I had figured/hoped that by the time IE7 rolled around, enough sites will have implemented it that the IE team couldn't simply break the relatively trivial hack without first fixing the much less trivial bugs it was working around.

However, it's been clear for some time that * html was going to be fixed. And now today, via WaSP, comes news that IE7 is going to be fixing a whole lot of other hacks. Wording on the IE Blog ("how easy it is to fall into the CSS hack trap") seems to indicate that IE developers are against hacks in general, so it's probably reasonable to assume that they'll be looking to fix any others in the forthcoming releases.

Of course this was irritating at first. My initial reaction is to wonder why they're throwing out the medicine without first curing the ailment. Full CSS2 support in IE7 is still a long way off.

But it's probably not as bad as all that. I ranted a few months ago about IE7 beta 1, and since that time the IE team has announced that IE7 beta 2 will see a pile of new fixes. A lot of the things we still need to work around in IE6 may potentially be fixed in IE7. It's anyone's guess as to whether the list of fixes on the IE Blog is final, but I'd be willing to wager we'll see even more updates by the time IE7 final is released. There's a good chance that in some cases, you won't have to change a line of code; the float bug your hack was fixing is no longer broken, so IE7 will simply render the initial rule properly, and not parse whatever corrective rules the hack piles on. IE6 will still behave as expected, and IE7 will treat the code as any other CSS-friendly browser. Hopefully. Maybe. In some cases.

Regardless, there's still a way out. The IE team prefers we use conditional comments, which will continue to function in IE7 and even offer specific targeting of that browser. They do exactly what CSS hacks do, in an officially-sanctioned way. The main problem with using them is simply that a lot of us aren't yet. In fact, I'd be happy to give up the hack habit forever if all other browsers were to implement conditional comments, or a similar concept. But given how often that idea gets abused (think querystrings or browser-specific Javascript) I suppose there's something to be said for moderation.

So, all those existing sites out there we've built that don't use conditional comments? Wait and see. IE7 beta 2 should be out in the next few months, we'll know more then.

Update: In the comments, "Agent Bishop" rightly points out that conditional comments do not work when you're running multiple versions of Internet Explorer on a single PC. As always, Position is Everything has you covered. (Is there anything they can't do?)

Still, it's a pretty contorted process, and I don't think it's reasonable to expect web developers to hack their registry just to test web sites. I'd like to see an officially-sanctioned way of making conditional comments work with multiple versions of Internet Explorer. IE Team, how about it?

Update 2: Eric Meyer points out that Dean Edwards' IE7 script could very well give us an easy way out of any transition headaches.

Permalink › | 33 comments

FontExplorer

October 12

Cross one off my to-do list: Linotype's new (free) FontExplorer X made the rounds last month when it originally launched, but I finally got a chance to install it the other day. On the off chance you use OS X and haven't heard about it yet, now you have. A Windows version is coming shortly. When it's available for both platforms, consider it a must-have tool.

Why would you need a font manager? Because you have hundreds or thousands of fonts installed on your computer. (If not, then obviously you don't need one.)

While probably not as true these days, there was a time when a large collection of type would eat up memory pretty quickly. Even with vast stores of RAM though, who wants to scroll through hundreds upon hundreds of fonts every time you want to select Zapfino? A font manager allows you to turn on and off at will only the fonts you need. A good one will also offer quick previews and customizable collections so that you can better organize your type.

FontExplorer has plenty of other goodies up its sleeve, too; it's basically iTunes for fonts, including things like ratings and smart lists. There's even a built-in font store. And oh did I mention it's completely free?

I've just started banging the tires, but already I can see why the rave reviews are pouring in. Here's one of my favourite pleasant discoveries: you can use the Comment field as a list of tags, and then later filter them with smart lists or searches. So you could create a list that includes all fonts except those tagged with UltraCondensed, Dingbat, and Ornamental, for example. One minor annoyance is that you can't add comments for entire families, only individual faces. That causes a bit of extra work in the short term, but it's no deal-breaker.

Linotype has done an excellent job with FontExplorer, and I can genuinely see it changing the way I use type. It can't hurt them that a free tool this good will only serve to inspire loyalty. I know I'll be throwing more of my font dollars their way as a direct result.

Permalink › | 11 comments

Format Change

October 11

Since many of you are likely reading this an RSS reader, I doubt this'll change a thing. For everyone else, I just made a quick change to this site. The full content of the past 7 articles are displayed on the home page. That's right, mezzoblue is a blog once again.

About a year or two ago, I decided to separate articles out into their own pages, and reduce the home page to a summary of the latest bits and pieces of what was recently added around here. It was appropriate at the time, but it's not working anymore. Along with that change came a subtle shift in my mindset; I was no longer writing quick posts, I was writing "articles". Any time I sat down to write, why, it had better be good and in-depth.

But that format is no fun anymore, as evidenced by the relative infrequency of updates around these parts. So it's gone. Let's see how this works. (And yes, that probably means at some point I'll be ditching the separated comments pages and merging them back into the individual post pages themselves.)

Permalink › | 22 comments

UPS Delivery

October 11

One for the "This is Broken" files.

In the tradition of Mark Hurst's This is Broken and 37 Signals' ongoing focus on customer service, there's something bugging me about UPS.

In the past year, I've moved both home and office. Keeping the addresses straight was tough for a while, but I think I've finally got a grasp on it. Occasionally though, I still get packages delivered to my home address. Naturally, I'm not there during the day.

Most of the time I can simply pass along the office address to those sending the packages, but sometimes it's not that easy. In those cases, after receiving the first delivery notice, I phone up UPS to redirect. (I never seem to remember the automated customer-service line shortcut [for most systems, press '0'], so I get stuck listening to their tediously comprehensive, non-skippable set of tips that cover everything from what to do if you missed the third delivery attempt to what action you can take if you accidentally ship your prize schooner-in-a-bottle to Mongolia.)

When I get through to an actual person and request the redirect, the first thing they'll ask for is my forwarding address, complete with postal code (US equivalent being the ZIP code). Given the moves, I now have two to remember and a few to try and forget; they continually blend together, and I always end up telling the rep to hold on while I Google my own damn postal code.

An aside on postal/ZIP codes: does anyone else get the feeling that these have outlived their usefulness? They exist for an extra level of precision when street naming is an inexact science at times, but isn't that a relatively easy problem that software should be solving these days? It seems to me that the using a postal code in 2005 is akin to exposing a unique database ID to a user; there should be better ways.

But that all brings me to the real problem—how come UPS doesn't remember my forwarding address from the last time I redirected? I asked last week whether this were possible, and received assurance that it was not. It's a pretty safe bet that every time I choose to redirect a package in the future, I'll be sending it to the same address. I really shouldn't have to give them the same address every single time; software can do that for me.

I don't necessarily mind having to take action and inform UPS of the redirect, since I can see a lot of ways the system could potentially be confused or abused were the redirect automatic. But it should take a lot less effort on my part to make it happen.

Keep in mind that I'm not the shipping party, and I didn't choose the shipping company; when I see obvious customer service holes in the response to what I'd assume is a common action, the next time I am the one making the choice I may just dial up the other guys first.

Permalink › | 17 comments